Server Data Batching for Backup Synchronization Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Servers with synchronization latency between main and backup data storage machines face errors when the main machine goes down, as actions are not immediately reflected on the backup, leading to potential data duplication and management complexities.
Innovation Solution
The method involves splitting data into disjoint batches with unique time slots, identifying and sending selected data within the current slot, and discarding the batch allocated to the time slot during an incident, ensuring that only the affected batch is discarded in case of an incident, thereby preventing data duplication and streamlining the backup process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If real time synchronization is implemented between main data storage machine and backup machine, then data consistency is improved, but system complexity and cost increase significantly
Solution Approach 1:
The data is divided into disjoint batches, each associated with a specific time slot. This segmentation allows the system to track and manage synchronization status of individual batches rather than requiring continuous real-time synchronization of the entire dataset, thereby reducing system complexity while maintaining data consistency.
Solution Approach 2:
The system pre-allocates time slots to batches before incidents occur. By establishing this scheduling framework in advance, the system can quickly determine which batches are affected by incidents without requiring complex real-time analysis, simplifying the synchronization management process.
2Reliability
If disjoint batches are discarded upon incident occurrence, then data duplication is prevented, but some data may be lost
Solution Approach 1:
The system extracts and identifies the specific batch associated with the incident time slot, discarding only that particular batch rather than all batches. This targeted approach prevents data duplication while minimizing data loss by preserving batches that were not affected by the incident.
Solution Approach 2:
The time slot acts as an intermediary that links incidents to specific batches. By using time slots as the connecting element, the system can precisely identify which batch to discard without affecting other batches, thereby maintaining data integrity while preventing duplication.
3Reliability
If time slots are set larger than maximum synchronization latency, then synchronization errors are reduced, but data transmission efficiency decreases
Solution Approach 1:
The system changes the time slot size parameter to be larger than the maximum synchronization latency. This parameter adjustment ensures that entire batches remain consistent throughout their time slots, reducing synchronization errors while the batched approach maintains overall transmission efficiency by processing data in manageable units.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The invention is a method sending a plurality of data (21, 22, 23) from a server (10) to a fleet of devices (11, 12, 13). The method comprises the following steps: - splitting said plurality of data (21, 22, 23) in a set of disjoint batches (31, 32), - uniquely allocating to each of said disjoint batches a time slot, - identifying a selected data belonging to the disjoint batch which is allocated to the current time slot and sending the selected data from the server (10) to one device of said fleet, - if an incident occurs at the server (10), discarding from the server (10) the disjoint batch allocated to the time slot during which the incident occurred.